Moving the insulin-regulated glucose transporter GLUT4 into and out of storage

The glucose transporter isoform GLUT4 is unique among the glucose transporter family of proteins in that, in resting cells, it is sequestered very efficiently in a storage compartment. In insulin-sensitive cells, such as fat and muscle, insulin stimulation leads to release of GLUT4 from this reservoir and its translocation to the plasma membrane. This process is crucial for the control of blood and tissue glucose levels. Investigations of the composition and structure of the GLUT4 storage compartment, together with the targeting motifs that direct GLUT4 to this compartment, have been extensive but have been controversial. Recent findings have now provided a clearer consensus of opinion on the mechanisms involved in the formation of this storage compartment. However, another controversy has now emerged, which is unresolved. This concerns the issue of whether the insulin-regulated step occurs at the level of release of GLUT4 from the storage compartment or at the level at which released vesicles fuse with the plasma membrane.     

Regulated transport of the glucose transporter GLUT4

In muscle and fat cells, insulin stimulates the delivery of the glucose transporter GLUT4 from an intracellular location to the cell surface, where it facilitates the reduction of plasma glucose levels. Understanding the molecular mechanisms that mediate this translocation event involves integrating our knowledge of two fundamental processes--the signal transduction pathways that are triggered when insulin binds to its receptor and the membrane transport events that need to be modified to divert GLUT4 from intracellular storage to an active plasma membrane shuttle service.     

Interaction of the Akt substrate, AS160, with the glucose transporter 4 vesicle marker protein, insulin-regulated aminopeptidase

Insulin-regulated aminopeptidase (IRAP), a marker of glucose transporter 4 (GLUT4) storage vesicles (GSVs), is the only protein known to traffic with GLUT4. In the basal state, GSVs are sequestered from the constitutively recycling endosomal system to an insulin-responsive, intracellular pool. Insulin induces a rapid translocation of GSVs to the cell surface from this pool, resulting in the incorporation of IRAP and GLUT4 into the plasma membrane. We sought to identify proteins that interact with IRAP to further understand this GSV trafficking process. This study describes our identification of a novel interaction between the amino terminus of IRAP and the Akt substrate, AS160 (Akt substrate of 160 kDa). The validity of this interaction was confirmed by coimmunoprecipitation of both overexpressed and endogenous proteins. Moreover, confocal microscopy demonstrated colocalization of these proteins. In addition, we demonstrate that the IRAP-binding domain of AS160 falls within its second phosphotyrosine-binding domain and the interaction is not regulated by AS160 phosphorylation. We hypothesize that AS160 is localized to GLUT4-containing vesicles via its interaction with IRAP where it inhibits the activity of Rab substrates in its vicinity, effectively tethering the vesicles intracellularly.     

Intracellular insulin-responsive glucose transporter (GLUT4) distribution but not insulin-stimulated GLUT4 exocytosis and recycling are microtubule dependent

To investigate the potential role of microtubules in the regulation of insulin-responsive glucose transporter (GLUT4) trafficking in adipocytes, we examined the effects of microtubule depolymerizing and stabilizing agents. In contrast to previous reports, disruption or stabilization of microtubule structures had no significant effect on insulin-stimulated GLUT4 translocation. However, consistent with a more recent study (Molero, J. C., J. P. Whitehead, T. Meerloo, and D. E. James, 2001, J Biol Chem 276:43829-43835) nocodazole did inhibit glucose uptake through a direct interaction with the transporter itself independent of the translocation process. In addition, the initial rate of GLUT4 endocytosis was not significantly affected by microtubule depolymerization. However, these internalized GLUT4 compartments are confined to regions just beneath the plasma membrane and were not exposed to the extracellular space. Furthermore, they were unable to undergo further sorting steps and trafficking to the perinuclear region. Nevertheless, these apparent early endocytic GLUT4 compartments fully responded to a second insulin stimulation with an identical extent of plasma membrane translocation. Together, these data demonstrate that although microtubular organization may play a role in the trafficking of GLUT4 early endocytic vesicles back to the perinuclear region, they do not have a significant role in insulin-stimulated GLUT4 exocytosis, initial endocytosis from the plasma membrane and/or recycling back to the plasma membrane.     

Isoproterenol inhibits cyclic AMP-mediated but not insulin-mediated translocation of the GLUT4 glucose transporter isoform

Isoproterenol is a beta adrenergic agonist whose effects have been attributed to the generation of cAMP. Previous studies have shown that it inhibits glucose transport in adipocytes without changing the number of insulin-responsive glucose transporters (GLUT4) on the cell surface. However, we have shown previously that cAMP stimulates translocation of GLUT4 to the cell surface in adipocytes (Keladaet al. J Biol Chem 267, 7021–7025, 1992). We therefore further investigated the mechanisms involved in isoproterenol regulation of glucose transport. Consistent with the effects of dibutyryl cAMP, we found that a low concentration of isoproterenol (10 nM) stimulated glucose transport and the translocation of GLUT4 from the low density microsomal fraction to the plasma membrane. By contrast, a higher concentration of isoproterenol (1 M) did not stimulate transport or GLUT4 translocation and furthermore inhibited dibutyryl cAMP-stimulated GLUT4 translocation. This inhibitory effect was specific for cAMP since isoproterenol had no effect on insulin-stimulated GLUT4 translocation. We conclude that isoproterenol has a biphasic effect on glucose transport, mediated by acute translocation of GLUT4 at low concentrations and by inhibition of intrinsic activity at high concentration, both of which may be explained by effects of cAMP. It has a further cAMP-independent effect at high concentration to inhibit cAMP-mediated translocation of GLUT4.This work forms portions of the PhD thesis requirements.     

Activation of the glucose transporter GLUT4 by insulin.

The transport of glucose into cells and tissues is a highly regulated process, mediated by a family of facilitative glucose transporters (GLUTs). Insulin-stimulated glucose uptake is primarily mediated by the transporter isoform GLUT4, which is predominantly expressed in mature skeletal muscle and fat tissues. Our recent work suggests that two separate pathways are initiated in response to insulin: (i) to recruit transporters to the cell surface from intracellular pools and (ii) to increase the intrinsic activity of the transporters. These pathways are differentially inhibited by wortmannin, demonstrating that the two pathways do not operate in series. Conversely, inhibitors of p38 mitogen-activated protein kinase (MAPK) imply that p38 MAPK is involved only in the regulation of the pathway leading to the insulin-stimulated activation of GLUT4. This review discusses the evidence for the divergence of GLUT4 translocation and activity and proposed mechanisms for the regulation of GLUT4.     

Golgin-160 is required for the Golgi membrane sorting of the insulin-responsive glucose transporter GLUT4 in adipocytes

The peripheral Golgi protein golgin-160 is induced during 3T3L1 adipogenesis and is primarily localized to the Golgi cisternae distinct from the trans-Golgi network (TGN) in a general distribution similar to p115. Small interfering RNA (siRNA)-mediated reduction in golgin-160 protein resulted in an increase accumulation of the insulin-responsive amino peptidase (IRAP) and the insulin-regulated glucose transporter (GLUT4) at the plasma membrane concomitant with enhanced glucose uptake in the basal state. The redistribution of GLUT4 was rescued by expression of a siRNA-resistant golgin-160 cDNA. The basal state accumulation of plasma membrane GLUT4 occurred due to an increased rate of exocytosis without any significant effect on the rate of endocytosis. This GLUT4 trafficking to the plasma membrane in the absence of golgin-160 was independent of TGN/Golgi sorting, because it was no longer inhibited by the expression of a dominant-interfering Golgi-localized, gamma-ear-containing ARF-binding protein mutant and displayed reduced binding to the lectin wheat germ agglutinin. Moreover, expression of the amino terminal head domain (amino acids 1-393) had no significant effect on the distribution or insulin-regulated trafficking of GLUT4 or IRAP. In contrast, expression of carboxyl alpha helical region (393-1498) inhibited insulin-stimulated GLUT4 and IRAP translocation, but it had no effect on the sorting of constitutive membrane trafficking proteins, the transferrin receptor, or vesicular stomatitis virus G protein. Together, these data demonstrate that golgin-160 plays an important role in directing insulin-regulated trafficking proteins toward the insulin-responsive compartment in adipocytes.     

Differential localization of glucose transporter isoforms in non-polarized mammalian cells: distribution of GLUT1 but not GLUT3 to detergent-resistant membrane domains

The hexose transporter family, which mediates a facilitated uptake in mammalian cells, consists of more than 10 members containing 12 membrane-spanning segments with a single N-glycosylation site. However, it remains unknown how these isoforms are functionally organized in the membrane domains. In this report, we describe a differential distribution of the glucose transporter isoforms GLUT1 and GLUT3 to detergent-resistant membrane domains (DRMs) in non-polarized mammalian cells. Whereas more than 80% of cellular proteins containing GLUT3 in HeLa cell lines was solubilized by a non-ionic detergent (either Triton X-100 or Lubrol WX) at 4 degrees C, GLUT1 remained insoluble together with the DRM-associated proteins, such as caveolin-1 and intestinal alkaline phosphatase (IAP). These DRM-associated proteins and the ganglioside GM1 were shown to float to the upper fractions when Triton X-100-solubilized cell extracts were centrifuged on a density gradient. In contrast, GLUT3 as well as most soluble proteins remained in the lower layers. Furthermore, perturbations of DRMs due to depletion of cholesterol by methyl-beta-cyclodextrin (m beta CD) rendered GLUT1 soluble in Triton X-100. Immunostaining patterns for these isoforms detected by confocal laser scanning microscopy in a living cell were also distinctive. These results suggest that in non-polarized mammalian cells, GLUT1 can be organized into a raft-like DRM domain but GLUT3 may distribute to fluid membrane domains. This differential distribution may occur irrespective of the N-glycosylation state or cell type.     

Mice with targeted disruption of spermidine/spermine N1-acetyltransferase gene maintain nearly normal tissue polyamine homeostasis but show signs of insulin resistance upon aging

The N(1)-acetylation of spermidine or spermine by spermidine/spermine N(1)-acetyltransferase (SSAT) is the ratecontrolling enzymatic step in the polyamine catabolism. We have now generated SSAT knockout (SSAT-KO) mice, which confirmed our earlier results with SSATdeficient embryonic stem (ES) cells showing only slightly affected polyamine homeostasis, mainly manifested as an elevated molar ratio of spermidine to spermine in most tissues indicating the indispensability of SSAT for the spermidine backconversion.Contrary to SSAT deficient ES cells, polyamine pools in SSAT-KO mice remained almost unchanged in response to N(1),N(11)-diethylnorspermine (DENSPM) treatment compared to a significant reduction of the polyamine pools in the wild-type animals and ES cells. Furthermore, SSATKO mice were more sensitive to the toxicity exerted by DENSPM in comparison with wild-type mice. The latter finding indicates that inducible SSAT plays an essential role in vivo in DENSPM treatmentevoked polyamine depletion, but a controversial role in toxicity of DENSPM. Surprisingly, liver polyamine pools were depleted similarly in wild-type and SSAT-KO mice in response to carbon tetrachloride treatment. Further characterization of SSAT knockout mice revealed insulin resistance at old age which supported the role of polyamine catabolism in glucose metabolism detected earlier with our SSAT overexpressing mice displaying enhanced basal metabolic rate, high insulin sensitivity and improved glucose tolerance. Therefore SSAT knockout mice might serve as a novel mouse model for type 2 diabetes.     

A role for phospholipase D in GLUT4 glucose transporter translocation

Based on recent studies showing that phospholipase D (PLD)1 is associated with intracellular membranes and promotes membrane budding from the trans-Golgi, we tested its possible role in the membrane trafficking of GLUT4 glucose transporters. Using immunofluorescence confocal microscopy, expressed Myc epitope-tagged PLD1 was found to associate with intracellular vesicular structures by a mechanism that requires its N-terminal pleckstrin homology domain. Partial co-localization with expressed GLUT4 fused to green fluorescent protein in both 3T3-L1 adipocytes and Chinese hamster ovary cells was evident. Furthermore, microinjection of purified PLD into cultured adipocytes markedly potentiated the effect of a submaximal concentration of insulin to stimulate GLUT4 translocation to cell surface membranes. Insulin stimulated PLD activity in cells expressing high levels of insulin receptors but no such insulin effect was detected in 3T3-L1 adipocytes. Taken together, these results are consistent with the hypothesis that PLD1 associated with GLUT4-containing membranes acts in a constitutive manner to promote the mechanism of GLUT4 translocation by insulin.     

Increased glucose transporter (GLUT4) protein expression in hyperthyroidism

We have studied skeletal muscle glucose uptake by perfused hindquarter preparations from rats treated with thyroxine. Basal glucose uptake (in the absence of insulin) was approximately 2 fold higher in muscle of hyperthyroid rats compared to controls. Insulin (10(-7) M) stimulated glucose uptake 4.0 and 6.8 fold in the 10 day and 30 day controls rats, respectively. Maximal glucose uptake (10(-7) M insulin) was not different in control and hyperthyroid rats and thus insulin responsiveness in the hyperthyroid animals was reduced to 2.5 fold stimulation. The abundance of the insulin-sensitive glucose transporter protein (muscle/fat, GLUT-4), measured by Western blot analysis using polyclonal antisera, was higher in skeletal muscle from both groups of hyperthyroid rats. These studies indicate that thyroid hormones increase basal glucose uptake in skeletal muscle and this is due, at least in part, to an increment of GLUT-4 isoform. Increased expression of muscle glucose transporter proteins may be responsible for the increased peripheral glucose utilization seen in hyperthyroidism.     

Role of SGK1 kinase in regulating glucose transport via glucose transporter GLUT4

Insulin stimulates glucose transport into muscle and fat cells by enhancing GLUT4 abundance in the plasma membrane through activation of phosphatidylinositol 3-kinase (PI3K). Protein kinase B (PKB) and PKCzeta are known PI3K downstream targets in the regulation of GLUT4. The serum- and glucocorticoid-inducible kinase SGK1 is similarly activated by insulin and capable to regulate cell surface expression of several metabolite transporters. In this study, we evaluated the putative role of SGK1 in the modulation of GLUT4. Coexpression of the kinase along with GLUT4 in Xenopus oocytes stimulated glucose transport. The enhanced GLUT4 activity was paralleled by increased transporter abundance in the plasma membrane. Disruption of the SGK1 phosphorylation site on GLUT4 ((S274A)GLUT4) abrogated the stimulating effect of SGK1. In summary, SGK1 promotes glucose transporter membrane abundance via GLUT4 phosphorylation at Ser274. Thus, SGK1 may contribute to the insulin and GLUT4-dependent regulation of cellular glucose uptake.     

Molecular machinery involved in the insulin-regulated fusion of GLUT4-containing vesicles with the plasma membrane (review).

The GLUT4 facilitative glucose transporter protein is primarily expressed in muscle and adipose tissue and accounts for the majority of post-prandial glucose uptake. In the basal or non-stimulated state, GLUT4 is localized to intracellular membrane compartments sequestered away from circulating glucose. However, in response to agonist stimulation, there is a marked redistribution of the GLUT4 protein to the cell surface membrane providing a transport route for the uptake of glucose. This GLUT4 translocation can be divided into four general steps: (i) GLUT4 vesicle trafficking out of the storage pool, (ii) docking just below the cell surface, (iii) priming via the interactions of the SNARE proteins present on the vesicular and plasma membranes, and (iv) fusion of the GLUT4 vesicle with the plasma membrane. This review focuses on recent advances made in identification and characterization of the molecular events and protein interactions involved in these steps of insulin-stimulated GLUT4 translocation.     

Substitution of leucine for tryptophan 412 does not abolish cytochalasin B labeling but markedly decreases the intrinsic activity of GLUT1 glucose transporter

GLUT1 glucose transporter cDNA was modified to introduce a single amino acid substitution of leucine for tryptophan 412, a putative cytochalasin B photo-affinity labeling site. Although the mutated transporter was expressed into plasma membranes of Chinese hamster ovary cells, glucose transport activity of the mutated transporter was observed to be only 15-30% of that of the wild-type GLUT1 when glucose transport activity was assessed by 2-deoxyglucose uptake at 0.1-10 mM concentrations. Analysis of glucose uptake kinetics depict that a mutation induced a 3-fold decrease in turnover number and a 2.5-fold increase in Km compared with the wild-type GLUT1. Importantly, cytochalasin B labeling was not abolished but decreased by 40%, and cytochalasin B binding was also decreased. In addition, the results obtained with side-specific glucose analogs suggested that the outer glucose binding site of the mutant appeared intact but the inner binding site was modulated. These results indicate 1) tryptophan 412 is not a cytochalasin B labeling site(s), although this residue is located in or close to the inner glucose binding site of the GLUT1 glucose transporter, 2) substitution of leucine for tryptophan 412 decreases the intrinsic activity of GLUT1 glucose transporter, which is definable as the turnover number/Km, to approximately 15% of that of the wild-type.     

Subcellular localization and trafficking of the GLUT4 glucose transporter isoform in insulin-responsive cells

The rate-limiting step in the uptake and metabolism of Dglucose by insulin target cells is thought to be glucose transport mediated by glucose transporters (primarily the GLUT4 isoform) localized to the plasma membrane. However, subcellular fractionation, photolabelling and immunocytochemical studies have shown that the pool of GLUT4 present in the plasma membrane is only one of many subcellular pools of this protein. GLUT4 has been found in occluded vesicles at the plasma membrane, clathrin-coated pits and vesicles, early endosomes, and tubulo-vesicular structures; the latter are analogous to known specialized secretory compartments. Tracking the movement of GLUT4 through these compartments, and defining the mechanism and site of action of insulin in stimulating this subcellular trafficking, are major topics of current investigation. Recent evidence focuses attention on the exocytosis of GLUT4 as the major site of insulin action. Increased exocytosis may be due to decreased retention of glucose transporters in an intracellular pool, or possibly to increased assembly of a vesicle docking and fusion complex. Although details are unknown, the presence in GLUT4 vesicles of a synaptobrevin homologue leads us to propose that a process analogous to that occurring in synaptic vesicle trafficking is involved in the assembly of GLUT4 vesicles into a form suitable for fusion with the plasma membrane. Evidence that the pathways of signalling from the insulin receptor and of GLUT4 vesicle exocytosis may converge at the level of the key signalling enzyme, phosphatidylinositol 3-kinase, is discussed.     

Subcellular compartmentalization and trafficking of the insulin-responsive glucose transporter, GLUT4.

Insulin increases glucose transport into cells of target tissues, primarily striated muscle and adipose. This is accomplished via the insulin-dependent translocation of the facilitative glucose transporter 4 (GLUT4) from intracellular storage sites to the plasma membrane. Insulin binds to the cell-surface insulin receptor and activates its intrinsic tyrosine kinase activity. The subsequent activation of phosphatidylinositol 3-kinase (PI 3-K) is well known to be necessary for the recruitment of GLUT4 to the cell surface. Both protein kinase B (PKB) and the atypical protein kinase C(lambda/zeta) (PKClambda/zeta) appear to function downstream of PI 3-K, but how these effectors influence GLUT4 translocation remains unknown. In addition, emerging evidence suggests that a second signaling cascade that functions independently of the PI 3-K pathway is also required for the insulin-dependent translocation of GLUT4. This second pathway involves the Rho-family GTP binding protein TC10, which functions within the specialized environment of lipid raft microdomains at the plasma membrane. Future work is necessary to identify the downstream effectors that link TC10, PKB, and PKClambda/zeta to GLUT4 translocation. Progress in this area will come from a better understanding of the compartmentalization of GLUT4 within the cell and of the mechanisms responsible for targeting the transporter to specialized insulin-responsive storage compartments. Furthermore, an understanding of how GLUT4 is retained within and released from these compartments will facilitate the identification of downstream signaling molecules that function proximal to the GLUT4 storage sites.